SMART DEER IRELAND
Welcome to SMARTDEER.
Welcome to SMARTDEER.
The SMARTDEER project - funded by DAFM and UCD - is a research program led by UCD researchers aimed at producing up-to-date distribution maps of deer species in Ireland. This has been achieved by combining historical deer data collected across the nation (2000 - 2023) with new digital tools (web survey and smartphone application) able to guarantee a steady flow of deer data thanks to the help of all stakeholders and citizens involved in the research program.
The SMARTDEER project - funded by DAFM and UCD - is a research program led by UCD researchers aimed at producing up-to-date distribution maps of deer species in Ireland. This has been achieved by combining historical deer data collected across the nation (2000 - 2023) with new digital tools (web survey and smartphone application) able to guarantee a steady flow of deer data thanks to the help of all stakeholders and citizens involved in the research program.
Our project has been developed to work with stakeholders, for stakeholders. We have been leading the first nationally-coordinated initiative of deer in Ireland by collecting and analyzing empirical data across the country that will help managers to take evidence-based decisions.
Our project has been developed to work with stakeholders, for stakeholders. We have been leading the first nationally-coordinated initiative of deer in Ireland by collecting and analyzing empirical data across the country that will help managers to take evidence-based decisions.
Neither the up-to-date precise distribution nor the population density of the key species of deer was currently known prior to SMARTDEER, and no national coordination in the collection of deer data existed as well. Recent advances with technologies such as smartphone applications or digital deer mapping surveys were not implemented for Irish deer prior to SMARTDEER, and our project aimed at filling these gaps by introducing tools and methodological approaches that will allow national deer monitoring on real time.
Neither the up-to-date precise distribution nor the population density of the key species of deer was currently known prior to SMARTDEER, and no national coordination in the collection of deer data existed as well. Recent advances with technologies such as smartphone applications or digital deer mapping surveys were not implemented for Irish deer prior to SMARTDEER, and our project aimed at filling these gaps by introducing tools and methodological approaches that will allow national deer monitoring on real time.
You can find below key findings of our project.
You can find below key findings of our project.
Key findings
Key findings
Deer distribution and relative density of the 3 most widespread deer species updated to year 2022. The colour gradient refers to the relative probability of species occurrence (rescaled from 0 to 1 to improved readability) [published in the international peer-reviewed journal Ecography by Morera-Pujol et al. 2023]. The map has a resolution of 5 x 5 km.
Deer distribution and relative density of the 3 most widespread deer species updated to year 2022. The colour gradient refers to the relative probability of species occurrence (rescaled from 0 to 1 to improved readability) [published in the international peer-reviewed journal Ecography by Morera-Pujol et al. 2023]. The map has a resolution of 5 x 5 km.
Deer distribution and relative density by Morera-Pujol et al. 2023 has been obtained by using recent advances with Bayesian spatial modelling and combining presence-absence (top-row of the plot below) and presence-only (bottom row) data collected in the last two decades. Presence-absence refers to data where either the presence or the absence of a deer species has been confirmed by systematic surveys. Whereas presence-only refers to actual observations of a deer species (different data sources and collection methodologies) without information on where it was absent.
Deer distribution and relative density by Morera-Pujol et al. 2023 has been obtained by using recent advances with Bayesian spatial modelling and combining presence-absence (top-row of the plot below) and presence-only (bottom row) data collected in the last two decades. Presence-absence refers to data where either the presence or the absence of a deer species has been confirmed by systematic surveys. Whereas presence-only refers to actual observations of a deer species (different data sources and collection methodologies) without information on where it was absent.
More recently, we examined NPWS culling returns collected over the last 2 decades and fitted Bayesian disaggregation models to rebuild the spatio-temporal dynamics of key deer populations in Ireland. Key findings are depicted below and full results are available in Murphy et al. 2023. Distributions are now available for years 2000, 2006, 2012, 2018 corresponding to those when CORINE land use data were available to explain deer distribution as a function of habitat change. The new 2023 CORINE will be available soon and so the up-to-date distribution of deer.
More recently, we examined NPWS culling returns collected over the last 2 decades and fitted Bayesian disaggregation models to rebuild the spatio-temporal dynamics of key deer populations in Ireland. Key findings are depicted below and full results are available in Murphy et al. 2023. Distributions are now available for years 2000, 2006, 2012, 2018 corresponding to those when CORINE land use data were available to explain deer distribution as a function of habitat change. The new 2023 CORINE will be available soon and so the up-to-date distribution of deer.
Please note that the numbers 0 to 6 (log scale) correspond to 0 to 400 sika deer within 5 by 5 km pixels. Distribution originates from the disaggregation of culling return data after accounting for the number of licensed hunters.
Please note that the numbers 0 to 6 (log scale) correspond to 0 to 400 sika deer within 5 by 5 km pixels. Distribution originates from the disaggregation of culling return data after accounting for the number of licensed hunters.
Please note that the numbers 0 to 4 (log scale) correspond to 0 to 55 red deer within 5 by 5 km pixels. Distribution originates from the disaggregation of culling return data after accounting for the number of licensed hunters.
Please note that the numbers 0 to 4 (log scale) correspond to 0 to 55 red deer within 5 by 5 km pixels. Distribution originates from the disaggregation of culling return data after accounting for the number of licensed hunters.
Please note that the numbers 0 to 7 (log scale) correspond to 0 to 1000 fallow deer within 5 by 5 km pixels. Distribution originates from the disaggregation of culling return data after accounting for the number of licensed hunters.
Please note that the numbers 0 to 7 (log scale) correspond to 0 to 1000 fallow deer within 5 by 5 km pixels. Distribution originates from the disaggregation of culling return data after accounting for the number of licensed hunters.
Below an example (for year 2018) of how the disaggregation models work sensu Murphy et al 2023. A (red deer), B (Sika deer) and C (fallow deer) 5 x 5 km distribution maps obtained after "disaggregating" D (red deer), E (sika deer), and F (fallow deer) culling return data from NPWS, respectively. Bayesian disaggregation predicts distribution data based on high-resolution environmental rasters.
Below an example (for year 2018) of how the disaggregation models work sensu Murphy et al 2023. A (red deer), B (Sika deer) and C (fallow deer) 5 x 5 km distribution maps obtained after "disaggregating" D (red deer), E (sika deer), and F (fallow deer) culling return data from NPWS, respectively. Bayesian disaggregation predicts distribution data based on high-resolution environmental rasters.
We also looked at deer damage in the sites monitored by the National Forest Inventory NFI and combined them with deer distribution and relative density from SMARTDEER to understand which species of deer are responsible for the different types of damage (e.g. frying, bark stripping, browing). This is only one of the several applications of SMARTDEER data aimed at improving deer monitoring and management. This work is under consideration for publication in an international peer-reviewed journal (Brock et al. 2023). Brock's work also included forest damage risk scenarios as a function of deer relative density and co-occurrence of multiple deer species.
We also looked at deer damage in the sites monitored by the National Forest Inventory NFI and combined them with deer distribution and relative density from SMARTDEER to understand which species of deer are responsible for the different types of damage (e.g. frying, bark stripping, browing). This is only one of the several applications of SMARTDEER data aimed at improving deer monitoring and management. This work is under consideration for publication in an international peer-reviewed journal (Brock et al. 2023). Brock's work also included forest damage risk scenarios as a function of deer relative density and co-occurrence of multiple deer species.
In the 3 maps above, the predicted damage for all 1,681 sampling stations across Ireland for: a) bark stripping damage, b) browsing damage, and c) fraying damage. Please note that key hotspots of deer damage have been recorded in the areas flagged by SMARTDEER as red deer, fallow deer, and sika deer hotspots of occurrence, respectively. The values of the colour gradient represent the predicted damage likelihood, with smaller and lighter yellow colours indicating low levels of predicted damage and larger, darker colours representing higher levels of predicted damage.
In the 3 maps above, the predicted damage for all 1,681 sampling stations across Ireland for: a) bark stripping damage, b) browsing damage, and c) fraying damage. Please note that key hotspots of deer damage have been recorded in the areas flagged by SMARTDEER as red deer, fallow deer, and sika deer hotspots of occurrence, respectively. The values of the colour gradient represent the predicted damage likelihood, with smaller and lighter yellow colours indicating low levels of predicted damage and larger, darker colours representing higher levels of predicted damage.
Our monitoring tools (and their future)
Our monitoring tools (and their future)
We have developed two tools that have been shown to be extremely efficient in capturing deer data across Ireland.
We have developed two tools that have been shown to be extremely efficient in capturing deer data across Ireland.
A smartphone application (SMARTDEER) for stakeholders to conduct systematic surveys for deer while outdoors and at home. Our smartphone application is capable of collecting deer presence and deer density data, culling returns and random sightings among the others. This is an efficient way to guarantee that deer data are collected throughout Ireland into the future.
A smartphone application (SMARTDEER) for stakeholders to conduct systematic surveys for deer while outdoors and at home. Our smartphone application is capable of collecting deer presence and deer density data, culling returns and random sightings among the others. This is an efficient way to guarantee that deer data are collected throughout Ireland into the future.
Our second tool is much simpler and even more efficient: we designed a National Mapping Survey using a web survey application developed for this project. Expert stakeholder knowledge can be used to generate distribution maps of known deer presence. Differently from the smartphone application, which is meant to be used regularly by deer stakeholders, the web survey application is aimed at collecting deer distribution data once a year. This tool has been particularly efficient when it comes to generate up to date distribution data.
Our second tool is much simpler and even more efficient: we designed a National Mapping Survey using a web survey application developed for this project. Expert stakeholder knowledge can be used to generate distribution maps of known deer presence. Differently from the smartphone application, which is meant to be used regularly by deer stakeholders, the web survey application is aimed at collecting deer distribution data once a year. This tool has been particularly efficient when it comes to generate up to date distribution data.
Both tools have been successfully tested and used during the SMARTDEER project. They are now available and it is up to government agencies to use them in the future to promote deer monitoring and gather data that are strongly needed to constantly update the status of our deer populations.
Both tools have been successfully tested and used during the SMARTDEER project. They are now available and it is up to government agencies to use them in the future to promote deer monitoring and gather data that are strongly needed to constantly update the status of our deer populations.
Funded by
Funded by
Led by
Led by
Simone Ciuti, UCD assistant professor of Wildlife Biology
Simone Ciuti, UCD assistant professor of Wildlife Biology
Last update: March 2023